Workplace interventions to improve well-being and reduce burnout for nurses, physicians and allied healthcare professionals: a systematic review

Healthcare workers experience higher rates of psychological distress and job burnout than many other sectors due to psychosocial hazards, including chronic occupational stress. These stressors contribute to presenteeism, anxiety, depression, and poorer patient care outcomes. The COVID-19 pandemic exacerbated mental health burdens, with marked increases in sickness absence for mental ill health. Contemporary occupational health frameworks (e.g., the Job Demands–Resources model and Watson’s Human Caring theory) emphasise enhancing job resources, holistic care, and self-care to buffer job strain and improve well-being. Workplace interventions can target different levels: individual (e.g., mindfulness, meditation, yoga, gratitude journaling), managerial, or organisational (e.g., workload reduction, autonomy, job crafting). Interventions are also classified by prevention level: primary (eliminate sources of stress), secondary (manage stress responses), and tertiary (treat stress-related conditions). Prior reviews frequently focused on specific techniques (mindfulness, yoga) or populations (physicians, general practitioners). A gap remained for a comprehensive synthesis of all intervention types across nurses, physicians, and allied health professionals, particularly incorporating recent evidence including the COVID-19 period. Therefore, this review aimed to identify and analyse positive outcome measures from workplace interventions designed to support well-being and reduce burnout among these groups since 2015.

Existing literature indicates individual-focused interventions, especially mindfulness-based practices (MBIs), commonly improve well-being indicators and reduce stress among healthcare workers. Reviews by Lomas et al. reported overall positive associations between MBIs and well-being but noted inconsistent study quality and few robust RCTs. Cocchiara et al. found yoga beneficial for managing stress and burnout. Klein et al. reported benefits of MBIs on burnout among health professionals. DeChant et al. highlighted organization-directed interventions for physician burnout, while Murray et al. focused on general practitioners’ psychological well-being interventions. The literature suggests organisational interventions (e.g., workload reduction, job redesign/job crafting) may offer proactive and sustained benefits but are less studied. The JD-R model and Human Caring theory provide theoretical bases for designing interventions that enhance resources and holistic care. Despite growing recognition of psychosocial hazards (e.g., Safe Work Australia’s 2022 codes), institutions often require stronger evidence for large-scale organisational changes, underscoring the need for comprehensive, high-quality evaluations across professions and intervention types.

This systematic review followed PRISMA guidelines (protocol validated with an institutional research librarian and PRISMA checklist used in reporting). Searches covered CINAHL, Embase, Emcare, Medline, and PsycInfo on 2 May 2022 and again on 5 October 2022, plus Google Scholar (first five pages). Inclusion criteria: studies involving allied health personnel, physicians, or nurses; workplace well-being interventions (individual or organisational) with improving well-being or reducing burnout as a primary focus; quantifiable pre- and post-intervention outcomes using validated well-being/burnout instruments; English language; published from 2015 onward. Mixed samples including patients were included if healthcare worker data were extractable. Exclusions: studies without confirmed healthcare worker participants; interventions not primarily aimed at well-being/burnout; suggested but not implemented interventions; studies prior to 2015; non-English; inaccessible full texts; and systematic reviews (though considered in discussion). Study selection: Titles/abstracts and full texts were independently screened by two reviewers (CC, JC), with disagreements resolved via reconciliation and third reviewer if needed. Data extraction: Conducted by CC and verified by EB, JC, SP, MT, recording study design, objectives, location, participants, sample size justification, setting/institutions, measures, intervention details (type, delivery, repeatability, randomisation/blinding), timelines, ethics, attrition, quantitative results, and outcomes. Quality assessment: Two reviewers (CC, JC) used the validated Medical Education Research Study Quality Instrument (MERSQI; max score 18) assessing design, institutions, response/measurement validity, data analysis, and outcomes. Meta-analysis was not feasible due to heterogeneity of designs, interventions, and outcomes; results were synthesised narratively and descriptively.

Screening identified 1663 records; after duplicates and quality assessment, 33 studies were included (published 2015–2022), predominantly from the USA (n=22), with others from Australia (n=2), the Netherlands (n=2), and single studies from Ireland, Italy, Portugal, Brazil, Hong Kong, Japan, and Iran. Participant groups included nurses only (n=16), physicians only (n=5), both physicians and nurses (n=1), mixed healthcare workers including allied health (n=5), and general healthcare staff (n=6). Sample sizes ranged from 9 to 228. Eleven studies reported power analyses. Designs: RCTs (n=8), quasi-experimental (n=6), single-group pretest/post-test (n=19). Of 14 studies with controls, 3 used active controls, 4 no-treatment controls, and 7 waitlist controls. Interventions: 30 were individually focused; 3 organisationally focused. Prevention levels: 31 secondary-level, 2 primary-level. Mindfulness-based practices (MBE) featured in 20 studies; others included meditation, yoga, acupuncture, massage chairs, gratitude journaling, choir singing, professional coaching, job crafting, workload reduction, and peer support networks. Attrition was generally low: 19 studies reported 0% dropout; five reported ≥25% attrition. Durations ranged from a single 90-minute session to 6 months; follow-ups (1–12 months) were variably included. Organisational interventions (n=3): - Workload reduction (Gregory et al.): Quasi-experimental with control; significantly reduced emotional exhaustion (MBI) versus control; reductions sustained at 6 months; some outcomes (e.g., workload) trended back toward baseline at follow-up. - Job crafting (Gordon et al.): Quasi-experimental with waitlist control; significant reductions in emotional exhaustion (OLBI) and improved work engagement (UWES) in both general and individual job crafting groups vs control. - Peer support network (Wahl et al.): Single-group design; non-significant reduction in ProQOL-5 burnout; significant improvement in Compassion Satisfaction (CPI), with discordant CS changes on ProQOL-5 versus CPI (CPI possibly more sensitive). Individually focused interventions (n=30): - MBEs (n=20) via in-person, hybrid, or online delivery typically yielded significant improvements in perceived stress (PSS, DASS-21 Stress), emotional exhaustion (MBI EE), mindfulness (MAAS/FFMQ/FMI), and resilience (BRS, CD-RISC). Several reported sustained benefits at 3-month follow-up (e.g., Schroeder et al., Werneburg et al., Mistretta et al.). Notable findings include: reductions in EE and DP (e.g., Schroeder et al.; Rodrigues et al.), improved WHO-5 well-being (e.g., Mistretta et al.), and increased personal accomplishment (e.g., Ofei-Dodoo et al.). Some studies showed limited or no significant changes (e.g., Bianchini & Copeland; certain hybrid/yoga-integrated MBEs on burnout). - Online/web-based MBEs (e.g., BREATHE): Significant reductions in nurse stress (NSS) across RCT and pre-post designs. Smartphone-based resilience training produced smaller, less sustained effects than in-person MBE. - Meditation (n=2): Significant reductions in burnout (MBI/ProQOL-5), perceived stress (PSS), depression (BDI), and anxiety (STAI), with maintenance at 6-week follow-up in one study; resilience (CD-RISC) improved (Bonamer & Aquino-Russell). - Yoga (n=1): Significant improvements in self-care (HPLP II), mindfulness (FMI), and reductions in EE and DP. - Acupuncture (n=1): Significant reductions in state/trait anxiety (STAI) and increased work engagement (UWES). - Massage chair (n=1): Reductions in perceived stress, heart rate, and blood pressure. - Gratitude journaling (n=3): RCT and pre-post studies showed significant decreases in perceived stress (PSS) and depressive symptoms (CES-D) and increases in subjective happiness (SHS); effects sustained to 3 months in some cohorts. - Choir singing (n=1): No significant quantitative changes; qualitative gains in social connectedness, enjoyment, and engagement. - Professional coaching (n=1): RCT showed a 19.5% decrease in high EE prevalence in the intervention group vs a 9% increase in controls at 5 months; improved quality of life and resilience. Overall, 29/33 studies reported effective outcomes, including significant improvements in well-being, engagement, quality of life, resilience, and reductions in burnout, perceived stress, anxiety, and depression. However, heterogeneity of methods and limited follow-up constrained synthesis and generalisability.

This review addressed the question of which workplace interventions improve well-being and reduce burnout among nurses, physicians, and allied health professionals. Evidence since 2015 indicates that a wide range of interventions—especially mindfulness-based education and related relaxation strategies—can reduce stress and burnout indices and enhance well-being, resilience, and engagement. Organisational interventions showed promising, potentially more sustainable effects on emotional exhaustion and engagement, aligning with theory (JD-R) that resource-enhancing and stressor-reducing changes can buffer strain. However, organisational studies were under-represented and often lacked robust controls or long-term assessments, limiting definitive conclusions. The predominance of secondary-level, individual-focused interventions reflects feasibility and low disruption to clinical workflow, supporting near-term benefits but not necessarily addressing root causes of occupational stressors. Methodological issues—heterogeneous designs and measures, frequent use of waitlist controls (which may overestimate effects), small sample sizes, lack of active comparators, and limited follow-up—temper confidence in effect durability and comparative efficacy. Nevertheless, the consistency of positive short-term outcomes across multiple modalities suggests these interventions can form part of multifaceted well-being strategies. Future work should prioritise rigorously designed RCTs, especially at the organisational level, with active controls, standardised outcome measures, and extended follow-up to evaluate sustainability and system-level impact.

Workplace well-being interventions benefit healthcare workers by improving well-being, engagement, quality of life, mindfulness, and resilience, and by reducing burnout, perceived stress, anxiety, and depressive symptoms. Most included studies targeted secondary, individual-level stress management techniques (e.g., mindfulness, meditation, gratitude), likely due to feasibility. Organisational interventions (e.g., workload reduction, job crafting) also showed positive effects but were few. The main contributions of this review are its cross-profession, cross-intervention synthesis since 2015 and identification of effective strategies alongside methodological gaps. Future research should emphasise proactive, primary organisational interventions; robust RCT designs with appropriate (preferably active) controls; standardised, validated outcome measures; and long-term follow-up to assess durability and real-world implementation.

- Marked heterogeneity in study designs, interventions, outcome measures, and analytic approaches prevented quantitative meta-analysis and limited cross-study comparisons. - Organisationally focused interventions were under-represented (n=3), limiting conclusions about system-level strategies. - Many studies lacked control groups or used waitlist controls, which may inflate effect estimates; active comparators were uncommon. - Post-intervention follow-up was often absent or short-term; sustained effects are uncertain. - Variable sample sizes, with some small cohorts, and inconsistent reporting (e.g., missing mean scores in some studies) reduce generalisability. - The review excluded potentially relevant organisational studies that did not use validated well-being measures or did not prioritise well-being as a primary outcome. - Diverse participant groups and contexts (professions, settings, durations) complicate generalisation across healthcare environments.